Type X Pottery) Morobe Province) Papua New Guinea: Petrography and

Possible Micronesian Relationships

JIM SPECHT, IAN LILLEY, AND WILLIAM R. DICKINSON

THE STUDY OF PREHISTORIC INTERACTION BETWEEN ISLANDS AND ARCHIPEL­

agoes of the Pacific has been largely concerned with processes of colonization and the development of exchange networks, both of which involved a complex flow of people, goods, knowledge, languages and genes. As Gosden and Pavlides (1994: 163) point out, however, this does not mean that "Pacific societies ... were in contact over vast distances all the time," and there must have been occa­sions when interaction was not planned, predictable, or sustained, nor did it in­volve the large-scale relocation of people. Such contacts no doubt contributed to the complex archaeological and ethnographic picture in many areas, but some might have left little or no expression in the archaeological record (cf. Rainbird 2004: 246; Spriggs 1997: 190). We discuss here a possible example of this on Huon Peninsula on the north coast of New Guinea, where aspects of a prehistoric pottery known as Type X suggest contact between the peninsula and the Palau Islands of western Micronesia about 1000 years ago. Throughout the article, we use the term "Micronesia" solely in a geographical sense, without cultural impli­cations (cf. Rainbird 2004).

Prehistoric links involving both colonization and the transfer of technologies between the island groups of Melanesia-West Polynesia and various parts of cen­tral and eastern Micronesia seem well established through the evidence of linguis­tics (e.g., Bayard 1976; Blust 1986; Pawley 1967; Shutler and Marck 1975: 101), archaeology (e.g., Athens 1990a:29, 1990b:173, 1995:268; Ayres 1990:191, 203; Intoh 1996,1997,1999), biological anthropology (e.g., Swindler and Weis­ler 2000; Weisler and Swindler 2002), and cultural practices such as kava drinking (Crowley 1994). In contrast, Yap, Palau, and the Marianas of western Micronesia were settled from the west, after which they maintained links with eastern Indo­nesia and the Philippines as well as with communities to their east (Intoh 1997;

Jim Specht is a research fellow in anthropology at the Australian Museum, and honorary associate in archaeology, University of Sydney. Ian Lilley is a reader in Aboriginal and Torres Strait Islander Stu­dies at the University of Queensland. William R. Dickinson is a professor emeritus in the Depart­ment of Geosciences, University of Arizona, Tucson.

AsiaJl Pcrspcrfillc5, VoL 45, No.1 © 2006 by the University of Hawai'i Press.

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 25

Rainbird 2004: 81-82, 92-93, 147; Wickler 2004). Some genetic data and rock art indicate possible relationships between Palau and the New Guinea region to the south (Lum and Cann 2000; Rainbird 2004: 146), but by and large the New Guinea area has played only a minor role in discussions about prehistoric interre­gional contacts with various parts of Micronesia. To a large degree, this reflects the paucity of archaeological data in key areas, but the evidence from Type X and other sources summarized below suggests a complex history of interaction.

The possibility of a link between Type X and Micronesian pottery was first raised by Lilley (1988a: 94), who observed a superficial similarity between Type X and Yap Laminated ware in J\1icronesia (Gifford and Gifford 1959: 180-182). This comparison is no longer tenable on several grounds. We report here petro­graphic analyses of Type X that confirm the use of grog temper (crushed sherds or fired clay), a technological trait that does not occur in modern pottery of the New Guinea region (May and Tuckson 1982; Rye 1976: 116) but is known in recent and prehistoric pottery of Palau, Yap, and Pohnpei (Athens 1990a; Ayres 1990; Dickinson and Shutler 2000; Intoh and Dickinson 2002; Fitzpatrick et al. 2003; Osborne 1966). Comparisons of rim forms link Type X with prehistoric Palauan pottery rather than with that of Yap or Pohnpei.

TYPE X

Type X is one of five pottery styles on coastal Huon Peninsula and the Siassi Islands, Morobe Province, Papua New Guinea (Fig. 1) that form a discontinuous sequence from Lapita to the modern industries of Sio-Gitua-Sialum and Madang (Lilley 1988a, 1988b, 1991, 2002, 2004). Lilley (1988a) initially defined Type X

o

r()tAroP 0 ~ u\:)mboi o.~.",

f2~ 0/>., Cf C/)

S' <i'oi> M I ' 10 <5'1) aal. 1i ,..~. • uam

Gitua ~I' Siassi Is

Sialum

Huon Peninsula

Finschhafen;..-______ --------I • Tami Is Wuvulu Ma~s Is

Tumleo

Huon Gulf

o 50 100

km

144'

Fig, 1. Map of Huon Peninsula and New Britain, Papua New Guinea, showing localities mentioned in the text.

26 ASIAN PERSPECTIVES 45(1) SPRING 2006

from sherds excavated at Sio and on Tuam and Malai Islands in the Siassi group, but it is now known from surface collections and excavations at 32 open sites and rock shelters on coastal Huon Peninsula and the nearby islands, including New Britain (Table 1). Although burials of unknown age also occur in some

TABLE I. DISTRIBUTION OF TYPE X SITES ON HUON PENINSULA, THE SIASSI ISLANDS,

AND NEW BRITAIN, PAPUA NEW GUINEA

SITE LOCATION TYPE EXC'/SURF. TYPE X OTHER TOTAL

Huon Pen. (22) Finschhafen No code Tami Islands Open Surf. 5 117 122 No code ]abim Mission Open Surf. 3? nla nla KAM Timbulim Open Surf. 43 13 56 No code Peninsula Open Surf. 8 9 Wandokai/Kanomi KIC Wandokai Shelter Exc. 15 16 KID Wandokai Shelter Exc. nla nla nla Paradise Springs Kanomi Open Exc. n/a n/a Sialum KCJ Sialum Open Surf. 6 4 10 KCM Sia Island Open Surf. 3 25 28 KCN Sialum Open Surf. 24 24 KCP Sialum Open Surf. 1 1 KCR Kwangum River Open Surf. 22 10 32 KCV Sialum Shelter Surf. 1 1 KCX Sialum Shelter Surf. 3 3 KDH Sialum Shelter Surf. 3 3 KD] Sialum Shelter Surf. 1 2 KDS Sialum Shelter Surf. 1 KDU Sialum Open Surf. 3 12 KDV Sialum Shelter Surf. 1 Gitua KBW Gitua Open Surf. 3 37 40 KBZ Gitua Open Surf. 21 108 129 Sio KBP Sigawa Island Open Exc. 7 >1000 >1000 KBQ Teliata Point Open Exc. 261 ~8239 ~8500

Vitiaz Strait (3) Arop Island ]CB NW coast Open Surf. 66 67 Siassi Islands KL] Malai Island Open Exc. 754 10,853 11,607 KLK Tuam Island Open Exc. 285 854 1139 New Britain (6) Kove Islands FCL Poi Island Open Exc. nla nla FPR Poi Island Open Exc. nla nla Arawe Islands FNY Pililo Island Open Exc. 76 2420 FOH Adwe Island Open Exc. >150 ~13,000 >13,150 FO] Kumbun Island Open Exc. 1+ >7000 >7000 Kandrian FLE Awakuo Shelter Exc. 4 4

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 27

rock shelters with Type X on Huon Peninsula, the presence of this pottery in midden contexts at open sites suggests that it served a domestic purpose and was not solely or primarily a funerary ware. Furthermore, its distribution across Vitiaz Strait may signal the start of connections between New Guinea and New Britain that later developed into the extensive trading network observed in historical times (Lilley 1988b). There is no precursor to Type X on Huon Peninsula.

Table 1 lists the frequencies of Type X sherds where known (n/a indicates lack of data), together with a summary of other pottery present at each site listed in the "Other" column. At more than half of the localities, Type X is represented by very small samples «10 sheds), though the frequencies listed do not necessar­ily indicate the quantity present at each site. The small numbers for the Sialum area represent selective surface collections, whereas at FLE near Kandrian the four Type X sherds were the only pottery recovered from the test pit. In excava­tions at KLJ, KLK, and KBQ, Type X represented between 2 and 25 percent of all sherds recovered (Lilley 1986: Figs. 8.4, 8.9, 8.28), and in most excavation units it was a minor component of the pottery present. At FOH in the Arawes, Type X sherds were more common on the ground surface than the excavated sample would indicate, and the listed frequency is an estimate for both surface and excavated sherds. While there are clearly problems with estimating the abso­lute and relative frequencies of Type X at many sites, the overall impression is that it is not common at most localities.

Lilley (1988a) originally proposed that Type X began at c. 1600-1500 cal. B.P.

and coexisted with Ancestral Sio and Ancestral Madang pottery for several cen­turies, with Type X ending around 850-500 cal. B.P. (Lilley 1988a, 1988b, 2004: 90-91). Reconsideration of the dating evidence now places Type X between about 1000 and 500 cal. B.P. (Lilley and Specht in review).

Type X vessels are typically rounded-bottom pots with two forms of rims (Class 1 and Class 2) and occasionally carinated shoulders (Figs. 2, 4; Lilley 1988a). The exterior surfaces are often shiny with a slightly greasy feeling and range in color from red to reddish-brown or yellowish-red (Munsell 2.5YR 4/4, 4/6, 5/6; 5YR 4/4, 4/6, 5/6). This color is probably the result of the application of a slip. Some exteriors have been burnished all over or in irregular lines. The vessels were roughly finished, leaving signs of construction by an additive process, and many sherds have a tendency to fracture into platelets or irregular lumps. The Class 1 rims have internal and external flanges added to form a T shape; some of these sagged before firing as they were too heavy to be supported by the vessel body (Lilley 1988a: Fig. 2). Class 2 rims have nearly parallel sides without flanges. Dec­oration consists of incised lines in diagonal or criss-cross patterns, or groups of shell-edge impressions (Figs. 2, 4; Lilley 1988a: Fig. 3). Lip modifications include sharp notches, rounded indentations, and shell-edge impressions.

Drawing on petrographic work by Watchman (1986), Lilley described Type X as lacking obvious nonplastic inclusions, with many having a "laminated" appearance and a tendency to break into thin, flat platelets. Later appraisal of sherds recovered on Huon Peninsula by Specht in 1969 and by Specht and J. Kamminga in 1972, however, showed that some share the rim fonns and gen­eral appearance of Type X but contain obvious mineral and other inclusions. Type X thus appeared to be a more diverse group than Lilley's original definition allowed.

28 ASIAN PERSPECTIVES . 45(1) . SPRING 2006

PREVIOUS WORK ON THE COMPOSITION OF TYPE X

Watchman (1986; Lilley 1988a: 94) examined 17 Type X sherds within a larger sample of 134 sherds from sites KLJ, KLK, and KBQ and placed 14 Type X sherds in one paste group and the other three in a closely related one. He noted (1986: 516) the presence in one Type X sherd of rounded clay clasts which, he suggested, might be derived from "broken fragments of other pots." Six other sherds assigned to Ancestral Sio and Madang pottery also contained what ap­peared to be fragments of old sherds (Watchman 1986: 516-518, 526, 531-532). The Ancestral Sio and Madang sherds could be assigned with some confidence to clay sources similar or identical to those used by modern potters of the two areas. Elemental analysis by optical emission spectroscopy of three Type X sherds from Sio, Tuam, and the KIC shelter near Wandokai, as well as a clay sample from Tuam, did not rule out Tuam as a possible source for the sherds. Following a sug­gestion by J. Chappell, however, Lilley (1988a: 95-96) proposed that Type X could have been made on Huon Peninsula, possibly in the Sialum-Wandokai area.

Summerhayes (2000: 175-192) analyzed the mineralogy of two Type X sherds from the FNY and FOH Lapita sites in the Arawe Islands and found that they dif­fered from Lapita pottery and river sands on the south coast of New Britain. He concluded that the two sherds were "non-local to the south coast," but he did not suggest a possible source area (181).

RENEWED PETROGRAPHIC STUDY

Geological Framework

The northern coastal ranges of New Guinea are segments of a Paleogene island arc complex accreted tectonically to the New Guinea-Australia continental block. The latter was drawn into the evolving subduction zone of an arc-trench system that lay along the southern flank of an originally intraoceanic arc-trench system as it approached the continental block from the north and accreted progressively from west to east (Cullen and Pigott 1989; Pigram and Davies 1987). Ongoing subduction offshore along the axis of Huon Gulf and at the New Britain Trench along the northern margin of the Solomon Sea continues to draw Huon Penin­sula and New Britain toward the Papuan Peninsula (Abers and McCaffrey 1994; Silver et al. 1991).

High ranges in the interior of Huon Peninsula are underlain by Oligocene to Lower Miocene Finisterre Volcanics (Jaques 1976), underlying Eocene oceanic crust, and Middle Eocene to Lower Miocene volcaniclastic strata derived from erosion of the Paleogene Finisterre volcanic arc (Abbott et al. 1994). Pre-mid­Miocene volcanogenic assemblages of Huon Peninsula are comparable in age and lithology to the Paleogene basement of central New Britain (Dickinson 2000), which is a generic continuation of the same island arc complex along the strike to the east. Streams draining to the eastern shore of Huon Peninsula head mostly in exposures of post-mid-Miocene limestone that overlie and dip northeast off the uplifted volcanogenic assemblages of the high ranges (Abbott et al. 1994; Jaques and Robinson 1977).

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 29

The geologic framework of Huon Peninsula suggests that terrigenous mineral sands are probably rare along its coast, where calcareous sands derived from off­shore fringing reefs or eroded from limestone exposures inland are probably dom­inant. N oncalcareous sands are also unexpected in the Siassi Islands of emergent reef limestone, unless derived from pyroclastic blankets that could have been spread southward from nearby Umboi in the active Schouten volcanic arc.

Potential Huon-Siassi Tempers

Bedrock assemblages of both Paleogene and Neogene age from which sands could be derived along the Huon coast-or on Umboi and New Britain near the Siassi Islands-are thus all arc volcanic assemblages that should yield similar sands composed dominantly of plagioclase feldspar, augitic clinopyroxene, and volcanic lithic fragments (polycrystalline and polyminerallic rock fragments) present in varying proportions dependent upon local sedimentological relations. There is no geological basis upon which the source of generically different mainland and is­land sands can be inferred. Therefore, no fundamental differences in temper com­position can be expected, no matter where within its known distribution Type X was made, for local calcareous sands would also be indistinguishable petrographi­cally. Local clays present along the Huon coast and within the Siassi Islands are likely to derive from similar parentage in the limestone substratum and therefore to be of similar composition as well. Sherd petrography is thus a blunt tool for sourcing Type X. Nevertheless, the unusual petrography of Type X sherds offers hints concerning their origin.

The Sample

This consisted of 28 sherds from 12 sites between Sio and Finschhafen and on Tuam and Malai Islands (Table 2). Lilley submitted 17 thin-section slides pre­viously examined by Watchman, including 16 Type X sherds (4 from Tuam, 6 from Malai, and 6 from Sio). One sherd oflate prehistoric Madang pottery from Sio, previously assigned by Watchman to the Yabob source near Madang, was included for comparative purposes. Specht's sample (n = 11) was selected to pro­vide a wide geographical representation of Type X on Huon Peninsula. It con­sisted of sherds from the KAM open site near Finschhafen (1), two open and five cave sites at Sialum (8), and an open site at Gitua (2). These included sherds with obvious nonplastic inclusions and others without inclusions that stylistically fall within the range of Type X. Only one sherd was selected to represent KAM be­cause the sherds collected there appeared in hand specimen to show no composi­tional variation. For comparison with the sherd mineralogy, we also included a sample of mineral beach sand from Kulabi village, west of Sio on the north coast of Huon Peninsula.

The sample included one incised body sherd and a plain Class 1 rim, both from the KLK site on Tuam Island, and a Class 1 rim with shell-impressed design (KCN/16) and two incised Class 2 ri111$ (KDH, KDS/1C) from the Sialum area. These sherds and a selection of other Type X sherds that were not analyzed are illustrated in Figures 2 to 4.

30 ASIAN PERSPECTIVES 45(r) SPRING 2006

TABLE 2. RESULTS OF PETROGRAPHIC ANALYSES OF 27 TYPE X SHERDS AND ONE MADANG

STYLE SHERD FROM HUON PENINSULA AND THE SIASSI ISLANDS, PAPUA NEW GUINEA

AREA

Siassi Islands Tuam (4)

Malai (5)

SHERD CODE

KLK 5/1 (Ill/3) KLK 7/1 (Ill/3) KLK 4/2 (Ill/2) KLK 10/1 (Ill/3) KLJ 5/3 (1/1) KLJ 5/4 (1/1) KLJ 16/4 (I/2) KLJ 28/1 (1/4) KLJ 28/2 (1/4)

Huon Peninsula Finschhafen (1) KAM/33 Sialum (8) KCM/2

Gitua (2)

Sio (8)

Totals (N = 28)

KCN/16

KCN/18 KCR/3 KCV KCX/l KDH/5

KDS/IC

KBZ/2/13 KBZ/2/14 KBQ 6/4 (1/2)

KBQ 6/3 (1/2) KBQ 15/1 (1/2-3) KBQ 15/2 (I/2-3) KBQ 23/4 (1/3) KBQ 23/5 (1/3) KBQ 28/3 (1/4) KBQ 28/4 (1/4)

SHERD TYPE

body body, incised rim, class 1 body body body body body body

body body rim, class 1,

shell imp body body body body rim, class 2,

incised rim, class 2

incised body body body,

Madang body body body body body body body

Note: Excavation contexts in parenthesis where relevant.

GROG GROG GROG GROG FERRU-

A B C

x x

x

x

x x

x x

x

x x

x

x

x x

x

5 5 6

D GINOUS SAND

x

x x

x

x

x

x

x

x

8

x

x

x

3

Type X Sherd Petrography

In general, Type X sherds contain a paucity of terrigenous (silicate and oxide) mineral grains derived from volcanic bedrock, and only a few contain calcareous grains. Systematic use of coastal sands as temper is thus precluded, and in this re­spect the Type X tempers are distinctly atypical for Oceanian sherd suites (Dick­inson 1998; Dickinson and Shutler 2000).

Three broad groupings are identified, here termed grog-tempered, ferruginous­tempered, and sand-tempered. The three groups are gradational, without sharp divisions between them (Table 2). The distribution of the three gradationally

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 3 I

a b

r c --- -

d

o

Fig. 2. Type X rim sherds studied in thin-section by Dickinson (vessel interior to right). a: KLK 4/2-Tuam Island (grog temper A), KLK/III, layer 2; b: KCN/16-Sialum, shell-impressed (ferru­ginous temper), surface; c: KDH/I-5-Sialum, incised (grog temper B), mouth diameter 260 mm, surface; d: KDS/l-Sialum, incised (ferruginous temper), mouth diameter 180-200 mm, surface.

related sherd types is thus fully compatible with the inference that the sherds derive froITl the same place on Huon Peninsula, although that place could have been a finite stretch of coast rather than one specific locale.

Grog- Tempered Sherds - The main temper grains in a majority of the Type X sherds (n = 17) are grog particles, broken pottery of medium to coarse sand size, discernible only as irregular domains with internal color, texture, or fabric that contrast with the surrounding clay paste. The grog particles occasionally are rounded to subrounded, while others are angular. The following four subgroups are recognized:

1. Grog A: Five sherds contain almost exclusively grog temper: KCM/2, KCN/ 18, KCX/l, KLK 5/1, and KLJ 28/2.

32 ASIAN PERSPECTIVES . 45(1) . SPRING 2006

17 a b

9 h

1 '1"

T 1 1 c e

k

.~ ~.J "----===-=--- o p q

Fig. 3. Type X rims, all surface finds (orientations mostly unknown, vessel interior to right) . a: KCX/2-Sialum; b: KAM/53-Finschhafen; c: KCR/17-Sialum; d: KCR/5-9-19-Sialnm; e: KCR/1 2-Sialum; f: KAM/2 1-Finschhafen; g: KAM/14-Finschhafen; h: KAM/52-Finschhafen; i: FLE/ I/ l-Kandrian; j: KAM/19-Finschhafen; k: KAM/22-Finschhafen; I: KAM/ 18-Finschhafen; m: KCR/27-Sialum; n: KCR/13-Sialnm, incised; 0: KAM/ 29-Finschhafen; p: KAM/ 16-Finschhafen; q: KAM/15-Finschhafen, shell edge impressions on lip.

2. Grog B: Five sherds also contain subordinate ferruginous grains that serve to show generic continuity between the grog-tempered and ferruginous-tempered sherds of the Type X sherd suite: KAM/33, KDH/5, KBQ 6/3, KLK 7/ 1, and KLJ 5/3.

3. Grog C: Six sherds also contain subordinate terrigenous sand grains that document generic continuity between grog-tempered and sand-tempered sherds of the Type X sherd suite: KBQ 15/ 1, KBQ 23/4, KCV, KBZ/2/ 13, KLK 4/2, and KLJ 28/1 .

4. Grog D: One sherd (KLK 10/1) from Tuam contains a few calcareous sand grains that were probably reworked from consolidated limestone rather than derived from contemporaneous reef detritus.

Ferruginous- Tempered Sherds - The next most abundant tempering agent in Type X sherds (n = 8) are subspherical to ovoid particles of medium to coarse sand size, composed largely of semiopaque iron oxides and probably derived from fer­ruginous pedogenic nodules in soils derived from subjacent limestone exposures. Subordinate grog particles are present in all the ferruginous-tempered sherds, however, and show that the grog-tempered and ferruginous-tempered sherds are but end members of a spectrum of related sherds. Two ferruginous-tempered sherds (KBQ 23/5 and KLJ 5/4) also contain subordinate terrigenous sand grains that tie the ferruginous-tempered Type X sherds generically to the sand-

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 33

a

b

o Scm -=-=-

Scm -=-=-

Fig. 4. Type X forms and rims, all surface finds. a: KDU-Sialum, incised with lip notching, mouth diameter 200 mm; b: KDV-Sialum, burnished with lip notching, mouth diameter 260 mm; c: KAM/26-Finschhafen, rim with lug; d: KAM/13-Finschhafen, rim with lug.

tempered Type X sherds as well. One of the ferruginous-tempered sherds from Sio (KBQ 15/2) contains subordinate calcareous grains composed at least in part of skeletal bioclastic debris and probably derived from beach sand including a component of contemporaneous reef detritus. Ferruginous-tempered sherds lack-

34 ASIAN PERSPECTIVES . 45(1) . SPRING 2006

ing any appreciable admixture of sand grains are KBQ 28/4, KBZ/2/14, KCN/16, KDS/1C, and KLJ 16/4.

Sand- Tempered Sherds - Two Type X sherds (KCR/3 and KBQ 28/3) and the Madang Style sherd (KBQ 6/4) contain subangular to subrounded terrigenous grains of fine to medium sand, but the sand is consistently mixed with subordinate grog particles that serve to tie these sherds generically to the rest of the Type X sherd suite. There were no sand-tempered sherds in the Siassi sample. While this may provide presumptive support for the inference that Type X derives from the Huon coast, the presence in the Siassi sample of ferruginous-tempered sherds containing subordinate terrigenous sand of the same general type as the sand­tempered sherds may indicate that the lack of sand-temper in the Siassi sherds that were sectioned simply reflects the overall paucity of this temper in the total sample studied here.

The principal sand grain types present in the sand-tempered sherds include, in approximate order of relative abundance, clinopyroxene (augite), plagioclase (feldspar), and volcanic rock fragments including a variety of textural and com­positional types ranging from microlitic grains (with groundmasses of tiny plagio­clase microlites set in volcanic glass), in part microporphyritic, to grains of altered brown volcanic glass (mafic). Subordinate grain types include opaque iron oxides (probably dominantly magnetite) and rare orthopyroxene (hypersthene) and hornblende. The mineral grains are all types common as phenocrysts in volcanic rocks of the Vitiaz Strait region (Jaques 1976; Johnson 1976), and the sands are all volcanic sands probably derived from Paleogene bedrock of Huon Peninsula, although derivation from the Neogene volcanic chain of the Schouten arc cannot be ruled out on geological grounds. The pyroxene-rich character of the terrige­nous sands implies probable origin as local beach placers (Dickinson 1994).

Kulabi Beach Sand - This is a moderately sorted (fine to coarse) but strongly placered sand composed of subrounded to subangular grains of modestly hybrid character (terrigenous grains mixed with <10 percent calcareous grains of reef de­tritus). Table 3 compares the composition of the terrigenous fraction of the sand

TABLE 3. FREQUENCY PERCENTAGES OF GRAIN TYPES IN THREE SAND-TEMPERED SHERDS

AND IN THE TERRIGENOUS FRACTION OF KULABI BEACH SAND, HUON PENINSULA, PAPUA

NEW GUINEA, BASED ON AREAL COUNTS OF "N" GRAINS IN THIN SECTION

KeR/3 KBQ 28/3 KBQ 6/4 SHERD KULABI

GRAIN TYPE (TYPE x) (TYPE x) (MADANG) AVERAGE SAND

Plagioclase feldspar 25 12 11 16 0 Clinopyroxene 63 80 68 70 77 Orthopyroxene 1 1 1 1 Hornblende 1 1 trace 1 0 Olivine 0 0 0 0 Opaque iron oxides 3 2 6 4 Volcanic rock fragments 7 4 14 9 20

microlitic 3 2 6 4 17 glassy 4 2 8 5 3

Counts 123 669 653 515

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 35

with the less placered volcanic sands that occur as terrigenous temper mixed with grog particles in three Huon sherds. The respective sands are similar but not iden­tical. Contrasts in hornblende and olivine content may derive from the statistics of small numbers or operator variance and probably do not reflect any significant difference in the source rocks for the sands. The higher plagioclase content of the temper sands may reflect locally more intense placering for the Kulabi beach sand. The higher content of volcanic rock fragments in the Kulabi sand may reflect simply its coarser grain size, as these fragments tend to be larger in diameter than the clinopyroxene mineral grains.

Discussion

There is no obvious preferential spatial distribution between the sites of the vari­ants of the grog-tempered sherds or those with ferruginous temper. On the other hand, when the Type X localities are treated as four areal groups (Siassi, Finschha­fen, Sialum, and Sio-Gitua), there is a hint of patterning. Siassi is the only area where all four grog-tempers occur (nine sherds), and this reduces to three around Sialum (eight sherds) and two at Sio-Gitua (ten sherds). Grog B is the only tem­per group that occurs in all four areas, though several occur in three areas. The ferruginous and sand tempers and possibly some of the grog-tempers are probably absent from Finschhafen, where the Type X sherds display uniform composition in hand specimens and probably all belong to the Grog B group. Given the small samples from each area, however, sample bias cannot be ruled out and may ex­plain the absence of Grog A at Sio and Gitua and the presence of Grog D only on Tuam.

On present data, the tempers do not provide a guide to the place or places of Type X pottery manufacture, but are not inconsistent with a Huon Peninsula origin. As Type X pottery was transported widely around the Vitiaz Strait area, the sand temper in some sherds may also have been transported from unknown locales to the place or places of pottery making, although its composition is not obviously exotic. The sand temper in the protohistoric Madang sherd from Sio (KBQ 6/4), which shares with Type X the use of grog (though in reduced amounts), is not generically different from the sand tempers in Type X sherds from the Huon coast. This may indicate an origin on Huon Peninsula similar to that of Type X rather than the Madang area itself, though Watchman (1986:512, 531) assigned this sherd to the same source as a modern potter's clay mix at Yabob. On balance, the Kulabi beach sand is not a close match for the sherd tem­per sands, but their generic similarity does not preclude the conclusion that the sherd tempers may derive from sand collected somewhere on the coast of the Huon Peninsula.

The number of sherds is too small to suggest any patterning between temper type and aspects of form or decoration. The sample included four rims, two each of Class 1 and Class 2 (see Fig. 2). The Class 1 rim from Tuam (KLK 4/2) is assigned to the Grog C group and that from Sialum (KCN/16-shell impressed) to the Ferruginous group. The Class 2 rim sherds from Sialum are assigned to Grog B (KDH/5-incised) and the Ferruginous group (KDS l/C-incised). The relationship between the two rim classes is unknown, but the compositional vari­ation between these sherds might indicate different places of manufacture. The

ASIAN PERSPECTIVES . 45(1) . SPRING 2006

two incised sherds from Tuam and Sialum are both assigned to the Grog B group.

The independent recognition of grog temper by Watchman and Dickinson is the first confirmation of the use of this temper category in the New Guinea re­gion in either prehistoric or recent pottery (May and Tuckson 1982; Rye 1976). The nearest occurrence of the technique is in Palau, western Micronesia, where Osborne (1966: 33) recorded crushed sherds as a tempering agent in recent times. The grog particles in Type X might similarly have derived from sherds or from fired clay prepared specifically to produce grog particles when smashed.

An alternative possibility for the origin of the grog temper might be the use of natural materials that present the appearance of crushed sherds on firing, though testing of this would require extensive field studies and experimentation beyond the scope of our present study. May and Tuckson (1982: 98, 150, 152, 238, 310) recorded the use of two or more clays in several contemporary Papua New Guinea industries. Some potters on Tumleo Island, in the Aitape area of West Sepik Province, for example, use a clay mix consisting of three ingredients, one of which has a low clay mineral content and might be more appropriately termed a temper (310-311). This third ingredient is sun-dried, crushed, and dry­sieved before the fine fraction is added to the other two clays. Through the cour­tesy of John Terrell (Field Museum, Chicago), we have examined 50 sherds from the Aitape coast, including seven from Tumleo. This sherd suite contains terrigenous temper sands of complex mineralogy, notably including quartz grains, derived ultimately from the diverse igneous, metamorphic, and sedimentary rock assemblages of the nearby mountainous hinterland. Both natural and manually added tempers are present in the Aitape sherds, including those from Tumleo, but none contains grog or groglike particles.

The use of grog temper in the Type X pots contrasts with the contemporary pottery technologies of Huon Peninsula and the Madang area. The potters of Sio, Gitua, and Sialum on Huon Peninsula do not temper their clay, although those of Sio No.2 and Nambariwa near Sio mix two clays to produce their potting mix. Indeed, Sio, Gitua, and Sialum are the only pottery centers using the paddle-and­anvil technique in contemporary Papua New Guinea that do not add a temper (May and Tuckson 1982: 152). The Madang potters use several clays to which they add beach sands as temper (169-171).

Petrography does not reveal why only grog was added to some Type X clay bodies, whereas ferruginous grains or terrigenous sand as well as grog were added to others. It might reflect relative ease of access to ferruginous concentrates in or on residual soil profiles and to placer sand deposits at different sites of Type X pottery manufacture. Alternatively, variation between individual potters might have come into play. The fact that modern potters of Sio, Gitua, and Sialum find no need to add a temper suggests that its use in the past was probably culturally rather than technologically driven. It is unlikely to have resulted from the use of clay sources different to those used today, as Watchman (1986: 528) suggested that a Type X sherd with Grog C (KBQ 15/1) might have been made from a contemporary Sio clay source.

Additional work is clearly required on the issue of grog temper and Type X technology. Here we retain the term "grog" without defining its specific origin and note that this technology appears to have been a fundamental characteristic of Type X production, which may account for the unique nature of this pottery

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 37

within the region. Furthermore, the use of grog in some Ancestral Sio and Madang pottery as well as in Type X raises the possibility of a developmental or historical relationship between them.

A POSSIBLE MICRONESIAN ORIGIN FOR TYPE X

In his formulation of Type X, Lilley (1988a: 94) noted general visual similarities between Type X and Laminated pottery on Yap in Micronesia. At that time, Yap Laminated ware was insecurely dated (Intoh 1984), but more recent work now places its beginning around 650 cal. B.P., toward the end of Type X (Descantes 2002: 229; Intoh 1990, 1992: 75). Furthermore, the vessel shapes of Yap Lami­nated pottery are unlike those of Type X, and its distinctive laminated structure is probably the result of wetting the pots prior to firing (Intoh 1988: 121, 1990:47). The superficial similarity between Type X and this Yap pottery is therefore rejected, and our focus is on the use of grog temper.

Hitherto, grog temper in past or recent Pacific pottery industries has been recognized only in pottery made in Palau, Yap, and Pohnpei in Micronesia (Athens 1990a; Ayres 1990; Dickinson and Shutler 2000: 215; Fitzpatrick et al. 2003; Intoh and Dickinson 2002; Osborne 1966: 35-37). Dickinson (1993) once proposed its use in American Samoa but now regards this as unlikely. The use of grog temper in Micronesia started about 2400 years ago in Palau (Clark 2004: 27-28, in review; Intoh 1992:75) and around 2000 cal. B.P. on Yap (Intoh 1990:44) and Pohnpei (Ayres 1990: 190). It continued in Palau until recent times (Osborne 1966: 35-37) but ended in Yap around 600-650 cal. B.P. (Intoh 1990: 44) and in Pohnpei around 1000-750 cal. B.P. (Athens 1990a: 26; Ayres 1990: 190).

There is thus a long history of grog temper in Palau, Yap, and Pohnpei extending back at least one millennium before the appearance of Type X and overlapping with most of its known span. We have argued above that the use of grog temper in Type X was probably culturally rather than technologically driven and, as there is currently no likely antecedent for Type X on the north coast of New Guinea, we raise the possibility that the sharing of the unusual grog temper technology between Type X and the Palau, Yap, and Pohnpei industries might indicate contact between one or more of those northerly centers and the Huon Peninsula region. Of particular interest here is the presence in Palau of four grog and composite tempers forming "gradational types" within a "single temper spec­trum" (Fitzpatrick et al. 2003: 1179), a situation remarkably similar to that found with Type X tempers. In both areas, grog was one of several options for temper. We now turn to the evidence from vessel forms.

We noted above the lack of correspondence between Type X forms and those of Yap Laminated pottery, and this seems to hold for earlier Yapese pottery (Intoh 1988: Fig. 6; Intoh and Leach 1985: Figs. 30-32). Similarly, the pottery from Pohnpei, which ended just after Type X began, bears no resemblance to Type X (e.g., Athens 1990a; Ayres 1990). The closest similarities are between Type X and pottery from Palau. The Class 1 T -shaped rim forms of Type X (Lil­ley 1988a: 92) are similar to some Palauan forms (conLpare Figs. 2-4 with Clark in review: Fig. 4; Clark and Wright 2003: Fig. 3 upper and middle; Hayakawa 1979: Fig. 9; Intoh 1981: Fig. 3; Osborne 1966: Fig. 15; Takayama 1979: Figs. 4-5; Takayama et al. 1980: Figs. 7-10). On Ulong Island, the Palauan examples start around 1000 cal. B.P. (Clark in review: Table 1). This is consistent with dates of

ASIAN PERSPECTIVES . 45(1) . SPRING 2006

probably not older than 1150-850 cal. B.P. on Kayangel Atoll (Takayama et al. 1980: Table 1, N-3369) and about 750 cal. B.P. or later on Ngarenggol (Takayama 1979: 90). In each case the dates fall within the time range of Type X of about 1000 to 500 cal. B.P. There are thus similarities of both form and temper technology between Type X and Palauan pottery of the same period. We note, however, that lugged rims similar to those of Type X at KAM (Figs. 4c, 4d) ap­pear to be absent from Palau (for undated non-Type X examples on Huon Pen­insula, see Abramson 1969: Plate IIa for the Tami Islands; Keysser 1911: Fig. 64 for Jabim Mission). With this exception, we suggest that Type X might have had its origin in the Palau group.

The distance between Huon Peninsula and the Micronesian islands producing grog-tempered pottery is at least 2000 km. The western end of New Guinea, however, lies about 1200 km from Yap and about 700 km from the main islands of Palau. Nuclear Micronesian (Chuukic) languages are or were once spoken on the most southerly islands of the Palau group, which extend to within about 100 km of the western end of New Guinea (Marck 1986; Wurm and Hattori 1981: Map 18). That this short distance was crossed is indicated by the similarities be­tween the rock art of Palau and parts of New Guinea (Rainbird 2004: 49, 146). Within Micronesia, the transport of Yapese and Palauan grog-tempered pottery to other islands began about 1900 years ago, with pottery from both sources traveling up to 700 km (Descantes 2002; Intoh 1992: 80; Intoh and Dickinson 2002: 127). This widespread movement might have been tied up with the devel­opment of the sawei exchange system and/or the export of the limestone discs (stone money) from Palau to Yap around 1000-1450 years ago (Fitzpatrick 2002; Rainbird 2004: 160). With such a long history of sea travel within western Micronesia, deliberate or accidental voyages along the north coast of New Guinea and to the Bismarck Archipelago seem highly probable (cf. Marck 1986; Spriggs 1997 : 189-192). Contacts in the reverse direction, particularly from New Guinea to Palau, are also likely in light of mtDNA data (Lum and Cann 2000: 165-166).

In his study of Pacific exploration and colonization, Irwin (1992) did not pre­sent computer simulations of voyages from the Palau area, but his Figures 41 and 42 of wind patterns in January and July suggest that deliberate two-way voyages from the Palau group to New Guinea and the Bismarck Archipelago were possi­ble. Records of one-way accidental voyages support this. A drift voyage from Palau reached the Mortlock Islands to the northeast of Bougainville, a distance of about 2400 km, and five men survived a voyage of about the same distance from eastern Indonesia to the Witu Islands off the north coast of New Britain (Anon. 1962: voyages 148, 152). Assuming that the potters of Palau at the time of Type X were female, as are the contemporary potters of Palau (Osborne 1966: 32) and Sio-Gitua-Sialum and Madang (May and Tuckson 1982), women must have been involved in the contact that introduced Type X to Huon Peninsula. This inter­pretation is supported by the fact that many drift voyages included women, and a young female survivor on the Palau canoe that reached the Mortlocks married a local man.

OTHER POSSIBLE MICRONESIAN LINKS

There are other possible indicators of contacts between parts of Micronesia and the north coast of New Guinea, although these do not all involve Palau. The lin-

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 39

guistic evidence is limited but evocative. The area covered by Type X primarily has Austronesian languages of the North New Guinea (NNG) Linkage (Ross 1988:120-189; Ross et al. 2003: Map 4). Several of these NNG languages (Adzera, Sio, Gitua) have a reflex of POc *kuron for "cooking pot" (May and Tuckson 1982: 152; Osmond and Ross 1998: 68; Ross 1996a: 70). In contrast, others (e.g., Mapos-Buang and Patep deg "cooking pot"; Bilibil daig "dish") have reflexes of POc *d(r)ag(i) for "container," which "is apparently reflected only in Yapese [= coconut-shell container] and NNG languages," though it is not clear whether this is a retention or a borrowing (Osmond and Ross 1998: 75; gloss in square brackets added).

The Madang coast and the Manus group are the only parts of north New Guinea and the Bismarck Archipelago where kava drinking was practiced. Draw­ing on linguistic and botanical evidence, Crowley (1994; cf. Spriggs 1997: 191-192) suggests that kava drinking began in the Vanuatu area, whence it was taken via West Polynesia to Kosrae and Pohnpei and eventually to Manus and the Madang coast. There is no indication when this occurred (for an alternative view for Manus, see Ambrose 1991).

The Huon Gulf-Finschhafen area is one of the few parts of the New Guinea region where compound fishhooks with a shell lure shank occur (Anell 1955:145-147, Fig. 12 right; Neuhauss 1911 vol. 1: Fig. 190c-d; Nordhoff 1930). Such hooks are common in the Solomon Islands (Bell et al. 1986), the likely source for some compound fishhooks of Micronesia (Palau: Takayama 1979:96; Fais: Intoh 1991:11-12; Pohnpei: Ayres 1990:191; Majuro: Weisler 2000: 128-131). The recent Huon Gulf-Finschhafen forms differ from those of the Solomons and Micronesia, but a shank fragment with a knobbed line attach­ment and incised line decoration found in layer 2 of KBQ/I at Sio recalls early Micronesian examples (Lilley 1986:382, Fig. 9.16). This Sio example is dated to­ward the end of Type X at 700-540 and 660-430 cal. B.P. (ANU 4332 and ANU 4990 respectively; both samples are marine shell, calibrated by Calib. Rev. 4.4.2, with Delta R = 0 ± 0; Stuiver et al. 1998a, 1998b; Stuiver and Reimer 1993). A similar shank, made from Pinctada shell and with a knobbed line attachment and incised line decoration, was found with a Type X sherd in rock shelter KIC to the south of Sialum (Araho 1995: 31, Fig. 8). This level is undated, but an age similar to that at Sio seems reasonable. In Micronesia such shanks appeared some­what earlier at 1500-1200 cal. B.P. on Fais, and ",1200 cal. B.P. on Majuro, Pohn­pei, and Kosrae (Intoh 1999:415-416), though Ayers (1990:191) and Weisler (2000: 128) suggest slightly earlier dates for Pohnpei and Majuro. While a Solo­mon Islands source for the Huon Gulf-Finschhafen fishhooks cannot be ruled out, the presence of comparable forms in Kosrae and Pohnpei adds to the argu­ment for past contact between these islands and the north coast of New Guinea indicated by the kava evidence. Gillett (1987: 12) records that "very old Satawa­lese fishermen spoke of a reef near New Guinea" that was a source of Pinctada maxima shell for making fishing lures on Satawal to the east of the Palau group, but it is not clear from his account whether the importing of this shell to Satawal existed before the arrival of Japanese divers in the twentieth century.

We noted earlier the claim for similarities between rock art in the Palau group and that of the western end of New Guinea (Rainbird 2004: 145-146). There are also similarities between Palauan rock paintings and those of the Manus Islands in the Bismarck Archipelago and between engravings of the "enveloped

4 0 ASIAN PERSPECTIVES . 45(1) . SPRING 2006

cross" motif at Pohnpaid on Pohnpei and similar forms in Island Melanesia (Rain­bird 2002:144,2004:196; Wilson 2002:78-79). In the Vanuatu area, the "enveloped cross" is assigned to the Rectilinear rock art tradition, which began about 1000 years ago (Wilson 2002: 222), though an earlier date is possible (Rainbird and Wilson 2002: 636). At this stage, however, we see no specific simi­larities between Type X designs or the rock art of Huon Peninsula and rock art in Micronesia.

Other evidence for contact between Micronesia and the New Guinea region relates to the northern islands of the Bismarck Archipelago, where there are indi­cations of links between languages of the Manus area and those of Nuclear Micronesian and Y apese, though the case of Yapese itself is highly complex (Ross 1988: 326-329, 1996b). There are also the backstrap loom in the Mussau Islands (Parkinson 1999: 143, 148) and the "Micronesian" material culture of Aua and Wuvulu Islands in Manus Province (Hambruch 1908; Parkinson 1999: 183-197), although Rainbird (2004: 150) introduces a caution about the antiquity of this latter situation. In the reverse direction, an obsidian point fragment found on Pohnpei is said to be similar to those of Manus (Ayres and Mauricio 1987: Fig. 4; Rainbird 2004: 186; Spriggs 1997: 128). Finally, modern Manus obsidian has been reported in the Marshall Islands, most likely as a result of floating volcanic debris (Spennemann and Ambrose 1997).

CONCLUSIONS

Our study supports the view that Type X was probably made on Huon Peninsula, without identifying a specific place or region, and confirms the use of a range of grog and other tempers. The use of grog temper cannot be ascribed to the nature of the raw materials available, as the modern potters of Sio-Gitua-Sialum do not add temper to their clays, and thus a cultural reason for its use seems likely. Com­parisons with the pottery of Palau, Yap, and Pohnpei-the only other areas of the western Pacific where grog temper is known-lead to the tentative conclusion that grog-tempered pottery was introduced to Huon Peninsula through contacts with Palau that involved the transfer of potters between the two areas.

The diverse evidence for contacts between other parts of Micronesia and the New Guinea region lend support to this suggestion. In no instance is there need to invoke large-scale movements of people; rather, there were probably many small-scale interactions, perhaps as a result of unintentional voyaging at different times and from different points. A similar situation has been suggested for the ori­gin of pottery of Kosrae, Pohnpei, and Chuuk. Athens (1990a: 29, 1995: 268) accepts the general view of derivation from Late Lapita or Post-Lapita Transition pottery, but he broadens the possible source area beyond the traditional view of a source area in the southeast Solomons and northern Vanuatu and allows for dif­ferent islands to have been settled from different origin points within the Lapita distribution.

If our interpretation of the origins of Type X is sustainable, it may contribute to explaining the complex cultural picture of the Huon Peninsula-Huon Gulf region obtained through ethnographic and linguistic studies. The region has witnessed many small-scale population movements (Chowning 1986; Harding 1967: 12-13) that resulted in what Bradshaw (1997) has called a "population ka-

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 41

leidoscope," in which clear-cut cultural boundaries do not exist between com­munities (Bradshaw 2001; cf. Welsch and Terrell 1998 for another part of the north coast of New Guinea). The largest settlement within the distribution of Type X, outside the township of Finschhafen, is Sio on the northeastern corner of Huon Peninsula. Ostensibly a single ethnolinguistic community, Sio is in fact a community of many origins within which the "theme of Sio history is unity out of diversity" (Harding and Clark 1994: 36).

Parallel to this complexity of community histories is the constant movement of goods between different parts of Huon Peninsula and Huon Gulf, throughout which pottery was traded widely in recent times. In 1969 Specht saw pots from the Watut Valley, Sio-Gitua-Sialum, and Huon Gulf in Supang village, in­land from Finschhafen, and Lilley (1986: 67, Fig. 3.6) later recorded both Sio­Gitua-Sialum and Huon Gulf vessels in Wandokai village, between Sialum and Finschhafen on the Huon coast. The diversity of pottery styles represented in Abramson's 1969 collection from the Tami Islands reflects a similar situation of overlapping distributions in the period when the Tami Islands were a major cen­ter of trade. The possible origin of Type X that we have proposed here adds a further dimension to this complexity by introducing the vagaries of historical contingency. It may prove extremely difficult, at times perhaps impossible, to identify in the archaeological record the history of individual cultural elements, each of which is likely to have had "its own history, its own spatial and temporal distribution" (Terrell and Welsch 1997: 565, 568).

ACKNOWLEDGMENTS

Specht acknowledges funding from the Australian Museum, Ian Potter Foundation, and the Australian Research Council and its predecessors. Lilley acknowledges funding from the Australian National University, University of Queensland, and National Geographic Society. We thank the national government of Papua New Guinea and the Morobe and West New Britain Provincial governments for permis­sion to undertake the fieldwork and for logistical and other assistance. Sarah Ph ear, Geoff Clark, Michiko lntoh, Margaret Tuckson, and Malcolm Ross provided infor­mation and advice on a range of topics, and Matthew Spriggs, Marshall Weisler, Christophe Descantes, and Geoff Clark made helpful comments on versions of the article. None of these people, of course, have any responsibility for the outcome. John Terrell and two unnamed referees provided constructive comments on the submitted version and hopefully will find that the final result deals adequately with their concerns. Fiona Roberts (Fig. 1), Kevin Rains (Fig. 2a), and Ghada Daher (Figs. 2b-2d, 3, and 4) prepared the line drawings.

REFERENCES

ABBOTT, L. D., E. A. SILVER, AND J. GALEWSKY

1994 Structural evolution of a modern arc-continent collision in Papua New Guinea. Tectol1ics 13: 1007-1034.

ABERS, G. A., AND R. MCCAFFREY

1994 Active arc-continent collision: Earthquakes, gravity anomalies, and fault kinematics in the Huon-Finisterre collision zone, Papua New Guinea. Tectol1ics 13: 237-245.

ABRAMSON, J. 1969 A preliminary archaeological survey of the Tami Islands, Territory of New Guinea. Ar­

chaeology aHd Physical Al1thropology ill Oceal1ia 4(2) : 83-90.

42 ASIAN PERSPECTIVES . 45(1) SPRING 2006

AMBROSE, W.

1991 Manus, mortars and the kava connection, in Oceanic Culture History: Essays in Honour of Roger Green: 461-469, ed.]. Davidson, G. Irwin, B. F. Leach, A. Pawley, and D. Brown. Dunedin: New Zealand Journal of Archaeology.

ANELL, B.

1955

ANON

1962

Contribution to the History of Fishing in the Southern Seas. Studia Ethnographica Upsaliensia 9. Uppsala: University ofUppsala.

A Table of accidental and deliberate voyages in the South Pacific, in Polynesian Naviga­tion: A Symposium on Andrew Sharp'S Theory of Accidental Voyages: 137-153, ed.]. Golson. Wellington: Polynesian Society.

ARAHO, N. 1995 Results of the Excavation of Two Sea Cut Caves in the Holocene Reef Face at Wando­

kai, Finschhafen District, Morobe Province, Papua New Guinea. Unpub!. B.A. (Hons) thesis. University of Papua New Guinea, Waigani.

ATHENS,]. S.

1990a Nan Madol pottery, Pohnpei. Micronesica Supplement 2: 17-32. 1990b Kosrae pottery, clay and early settlement. Micronesica Supplement 2: 171-186. 1995 Landscape Archaeology: Prehistoric Settlement, Subsistence and Environment of Kosrae, Eastern

Caroline Islands, Micronesia. Honolulu: International Archaeological Research Institute.

AYRES, W. S.

1990 Pohnpei's position in Eastern Micronesian prehistory. Micronesica Supplement 2: 187-212.

AYRES, W. S., AND R. MAURICIO

1987 Stone adzes from Pohnpei, Micronesia. Archaeology in Oceania 22(1): 27-31.

BAYARD, D. 1976 The Cultural Relationships of the Polynesian Outliers. Studies in Prehistoric Anthropology 9.

Dunedin: Department of Anthropology, Otago University.

BELL, D.,]. SPECHT, AND D. HAIN

1986 Beyond the reef: Compound fishhooks in the Solomon Islands, in Traditional Fishing in the Pacific: Ethnographic and Archaeological Papers from the 15th Pacific Science Congress: 45-63, ed. A. Anderson. Pacific Anthropological Records 37. Honolulu: Bishop Museum.

BLUST, R. 1984 Malaita-Micronesian: An Eastern Oceanic subgroup? Journal of the Polynesian Society

93(2) : 99-140.

BRADSHAW, ].

1997 The population kaleidoscope: Another factor in the Melanesian diversity v. Polynesian homogeneity debate. Journal of the Polynesian Society 106(3): 222-249.

2001 Language, culture, and community boundaries around the Huon Gulf of New Guinea, in Archaeology, Language and History: Essays on Culture and Ethnicity: 283-306, ed.]. E. Ter­rell. Westport, CT: Bergin and Garvey.

CHOWNING, A.

1986 Refugees, traders, and other wanderers: The linguistic effects of population mixing in Melanesia, in FOCAL II: Papers from the Fourth International Conference on Austronesian Lin­guistics: 407-434, ed. P. Geraghty, L. Carrington, and S. A. Wurm. Pacific Linguistics C-94. Canberra: Department of Linguistics, Australian National University.

CLARK, G. 2004 Radiocarbon dates from the Ulong site in Palau and implications for Western Microne­

sian prehistory. Archaeology in Oceania 39(1): 26-33. In A 3000-year culture sequence from Palau, Western Micronesia. Submitted to Asian Per-review spectives.

CLARK, G., AND D. WRIGHT

2003 The colonisation of Palau: Preliminary results from Angaur and Ulong, in Pacific Archaeol­ogy: Assessments and Prospects: 85-94, ed. C. Sand. Les Cahiers de I'Archeologie en Nouvelle-Caledonie 15. Noumea: Department Archeologie, Service des Musees et du Patrimoine de Nouvelle-Caledonie.

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 43

CROWLEY, T. 1994 Proto who drank kava? in Austronesian Termil1ologies: COl1til1uity al1d Change: 87-100, ed.

A. Pawley and M. D. Ross. Pacific Linguistics C-127.Canberra: Department of Linguis­tics, Australian National University.

CULLEN, A. B., AND J. D. PIGOTT

1989 Post-Jurassic tectonic evolution of Papua New Guinea. Tectonophysics 162: 291-302.

DESCANTES, C. 2002 The history of exchange between Yap and Ulithi, Western Caroline Islands. Unpubl. pa­

per presented at the 17 th Congress of the Indo-Pacific Prehistory Association, Taipei, September 2002.

DICKINSON, W. R. 1993 Sand temper in prehistoric potsherds from the To'aga site, in The To 'aga Site: Three Mil­

Iml1ia ofPolYl1esian Occupatiol1 ill the Mal1u'a Islal1ds, American Samoa: 151-156, ed. P. V. Kirch and T. L. Hunt. Contribution 51. Berkeley: Archaeological Research Facility, University of California.

1994 Natural beach placer analogous to prehistoric island tempers. Joumal of the PolYl1esian Soci­ety 103(2): 217-219.

1998 Petrographic temper provinces of prehistoric pottery in Oceania. Records of the Australial1 Museum 50(3): 263-276.

2000 Petrography of sand tempers in prehistoric Watom sherds and comparison with temper suites of the Bismarck Archipelago. New Zealand Joumal of Archaeology 20 (1998): 161-182.

DICKINSON, W. R., AND R. SHUTLER JR.

2000 Implications of petrographic temper analysis for Oceanian prehistory. Joumal of World Pre­history 14: 203-266.

FITZPATRICK, S. M. 2002 A radiocarbon chronology of Yapese stone money quarries in Palau. Microl1esica

34(2): 227-242.

FITZPATRICK, S. M., W. R. DICKINSON, AND G. CLARK

2003 Ceramic petrography and cultural interaction in Palau, Micronesia. Journal of Archaeological Sciel1ce 30: 1175-1184.

GIFFORD, E. W., AND D. S. GIFFORD

1959 Archaeological excavations in Yap. Allthropological Records 18(2): 149-224.

GILLETT, R. 1987 Traditiol1al TUl1a Fishil1g: A Study at Satawal, Cel1tral Caroline Islal1ds. Bulletin in Anthro­

pology 1. Honolulu: Bishop Museum.

GOSDEN, C., AND C. PAVLIDES

1994 Are islands insular? Landscape vs. seascape in the case of the Arawe Islands, Papua New Guinea. Archaeology ill Oceania 29(3): 162-171.

HAMBRUCH, P.

1908 Wuvulu und Aua (Maty- und Durour-Inseln) auf Grund der Sammlung F. E. Hellwig aus denJahren 1902-1904. Mitteilul1gm aus dem Museumfur Viilkerkul1de, Hamburg 2(1).

HARDING, T. G.

1967 Voyagers of the Vitiaz Strait: A Study of a New Guil1ea Trade System. Monograph 44. Seattle: University of Washington Press.

HARDING, T. G., AND S. A. CLARK

1994 The Sio story of Male. Pacific Studies 17(4):29-51.

HAYAKAWA, S.

1979 Archaeological investigations of PAKR-3 in the Palaus, in Report: Cultural Allthropological Research 011 the Folk Culture ill the Westem Carolil1e Islal1ds of Microl1esia ill 1977: 73-78, ed. H. Kusakabe. Tokyo: Committee for Micronesian Research, Tokyo University of For­eign Studies.

INTOH, M. 1981 Reconnaissance archaeological work on Ngulu Atoll in the Western Caroline Islands.

Asial1 Perspectives 24(1): 69-80. 1984 Yapese pottery: Review and problems. The Outrigger 3(1): 2-3.

44 ASIAN PERSPECTIVES • 45(1) . SPRING 2006

1988 Changing Prehistoric Yapese Pottery Technology: A Case Study of Adaptive Transfor­mation. Unpub!. Ph.D. diss. University of Otago, Dunedin.

1990 Ceramic environment and technology: A case study in the Yap Islands in Micronesia. Man and Culture in Oceania 6: 35-52.

1991 Archaeological Research on Fais Island: Preliminary Report. Tokyo: Department of Interna­tional Cultural Relations, Hokkaido Tokai University.

1992 Pottery traditions in Micronesia, in Poterie Lapita et Peuplement: 67-82, ed. J.-c. Gali­paud. Noumea: ORSTOM.

1996 Multi-regional contacts of prehistoric Fais Islanders in Micronesia. Bulletin of the Indo­Pacific Prehistory Association 15: 111-117.

1997 Human dispersals in Micronesia. Anthropological Science 105(1): 15-28. 1999 Cultural contacts between Micronesia and Melanesia, in The Pacific from 5000 to 2000

B.P.: Colonisation and Transformations: 407-422, ed. J.-c. Galipaud and 1. Lilley. Noumea: Institute de Recherche pour Ie Developpement.

INTOH, M., AND W. R. DICKINSON

2002 Prehistoric pottery movements in western Micronesia: Technological and petrological study of potsherds from Fais Island, in Fifty Years in the Field: Essays in Honour of Richard Shutler Jr's Archaeological Career: 123-134, ed. S. Bedford, C. Sand, and D. Burley. Mono­graph 25. Auckland: New Zealand Archaeological Association.

INTOH, M., AND F. LEACH

1985 Archaeological Investigations in the Yap Islands, Micronesia. British Archaeological Reports, International Series 277.

IRWIN, G. 1992 The Prehistoric Exploration and Colonisation of the Pacific. Cambridge, UK: Cambridge Uni­

versity Press.

JAQUES, A. L. 1976 High-K2 0 island-arc volcanic rocks from the Finisterre and Adelbert Ranges, northern

Papua New Guinea. Geological Society of America Bulletin 87: 861-867.

JAQUES, A. L., AND G. P. ROBINSON

1977 The continent/island-arc collision in northern Papua New Guinea. BMR Journal of Aus­tralian Geology and Geophysics 2: 289-303.

JOHNSON, R. W. 1976 Late Cainozoic volcanism and plate tectonics at the southern margin of the Bismarck Sea,

Papua New Guinea, in Volcanism in Australasia: 101-116, ed. R. W. Johnson. Amster­dam: Elsevier.

KEYSSER, C. 1911 Aus Leben der Kaileute, in Deutsch-Neu-Guinea vo!' 3: 3-242, ed. R. Neuhauss. Berlin:

LILLEY, 1. 1986

1988a

1988b

1991

2002

2004

Dietrich Reimer A. G.

Prehistoric Exchange in the Vitiaz Strait, Papua New Guinea. Unpub!. Ph.D. diss. Aus­tralian National University, Canberra. Type X: Description and discussion of a prehistoric ceramic ware from northeastern Pa­pua New Guinea. Bulletin of the Indo-Pacific Prehistory Association 8: 90-100. Prehistoric exchange across the Vitiaz Strait, Papua New Guinea. Current Anthropology 29:513-516. Lapita and Post-Lapita developments in the Vitiaz Straits-West New Britain area. Bulletin of the Indo-Pacific Prehistory Association 11 : 313-322. Lapita and Type Y in the KLK site, Siassi, Papua New Guinea, in Fifty Years in the Field: Essays in Honour of Richard Shutler Jr's Archaeological Career: 79-90, ed. S. Bedford, C. Sand, and D. Burley. Monograph 25. Auckland: New Zealand Archaeological Association. Trade and culture history across the Vitiaz Strait, Papua New Guinea: The emerging Post-Lapita coastal sequence. Records of the Australian Museum, Supplement 29: 89-96.

LILLEY, 1., AND J. SPECHT

In Revised dating of Type X pottery, Papua New Guinea. Submitted to the Records of the review Australian Museum.

LUM, J. K., AND R. L. CANN

2000 mtDNA lineage analyses: Origins and migrations of Micronesians and Polynesians. Ameri­can Joumal of Physical Anthropology 113: 151-168.

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 45

MARCK,J. C. 1986 Micronesian dialects and the overnight voyage. Journal of the Polynesian Society 95(1) : 253-

258.

MAY, P., AND M. TUCKSON

1982 The Traditional Pottery of Papua New Guinea. Sydney: Bay Books.

NEUHAUSS, R.

1911 Deutsch-Neu-GuiI1ea. 3 vols. Berlin: Dietrich Reimer A. G.

NORDHOFF, C. 1930 Notes on the off-shore fishing of the Society Islands. Journal of the Polynesian Society

39(3): 221-262.

OSBORNE, D.

1966 The Archaeology of the Palau Islands. Bulletin 230. Honolulu: B. P. Bishop Museum.

OSMOND, M., AND M. Ross

1998 Household artefacts, in The Lexicon of Proto Oceanic: The Culture and Environment of Ances­tral Oceanic Society, ed. M. Ross, A. Pawley, and M. Osmond. Vol. 1 of Material Culture: 67 -114. Pacific Linguistics C-152. Canberra: Pacific Linguistics, Australian National University.

PARKINSON, R.

1999 Thirty Years in the South Seas. Translated by J. Dennison and ed. J. P. White. Bathurst: [1907] Crawford House Publishing, in association with Oceania Publications, University of

Sydney.

PAWLEY, A.

1967 The relationships of Polynesian Outliers languages. Journal of the Polynesian Society 76(3): 259-296.

PIGRAM, C. J., AND H. L. DAVIES

1987 Terranes and accretion history of the New Guinea orogen. BMRJournal of Australian Ge­ology and Geophysics 10: 193-211.

RAINBIRD, P.

2002 Pohnpei petroglyphs, communication and miscommunication. Bulletin of the Indo-Pacijic Prehistory Association 22: 141-146.

2004 The Archaeology of Micronesia. Cambridge, UK: Cambridge University Press.

RAINBIRD, P., AND M. WILSON

2002 Crossing the line: The enveloped cross in Pohnpei, Federated States of Micronesia. Antiq­uity 76: 635-636.

Ross, M. D.

1988 Proto Oceanic and the Austronesian Languages of Western Melanesia. Pacific Linguistics C-98. Canberra: Department of Linguistics, Australian National University.

1996a Pottery terms in Proto Oceanic, in Oceanic Culture History: Essays in Honour of Roger Green: 67-82, ed. J. Davidson, G. Irwin, B. F. Leach, A. Pawley, and D. Brown. Dune­din: New Zealand Journal of Archaeology.

1996b Is Yapese Oceanic? In Reconstruction, Classijication, Description: Festschriji in Honour of Isi­dore Dyen: 121-166, ed. B. N othofer. Hamburg: Abera.

Ross, M., A. PAWLEY, AND M. OSMOND

2003 The Lexicon of Proto Oceanic: The Cultllre and Environment of Ancestral Oceanic Society. Vol. 2 of The Physical Environment. Pacific Linguistics 545. Canberra: Department of Linguistics, Australian National University.

RYE, O. S. 1976 Keeping your temper under control: Materials and the manufacture of Papuan pottery.

Archaeology and Physical Anthropology in Oceania 9(2): 106-137.

SHUTLER, R., AND J. C. MARCK

1975 On the dispersal of the Austronesian horticulturalists. Archaeology and Physical Anthropology in Oceania 10(2): 81-113.

SILVER, E. A., L. D. ABBOTT, K. S. KIRCHOFF-STEIN, D. L. REED, AND B. BERNSTEIN-TAYLOR

1991 Collision propagation in Papua New Guinea and the Solomon Sea. Tectonics 10: 863-874.

ASIAN PERSPECTIVES . 45(1) . SPRING 2006

SPENNEMANN, D.H.R., AND W. R. AMBROSE 1997 Floating obsidian and its implications for the interpretation of Pacific prehistory. Antiquity

71:188-193.

SPRIGGS, M. 1997 The Island Melanesians. Oxford: Blackwell.

STUIVER, M., AND P.]. REIMER 1993 Extended 14C database and revised CALlB radiocarbon calibration program. Radiocarbon

35: 215-230.

STUIVER, M., P.]. REIMER, E. BARD,]' W. BECK, G. S. BURR, K. A. HUGHEN, B. KROMER, F. G. MCCORMAC, ].V.D. PLICHT, AND M. SPURK

1998a INTCAL98 radiocarbon age calibration 24,000-0 cal. B.P. Radiocarbon 40: 1041-1083.

STUIVER, M., P.]. REIMER, AND T. F. BRAZIUNAS 1998b High-precision radiocarbon age calibration for terrestrial and marine samples. Radiocarbon

40: 1127-1154.

SUMMERHAYES, G. 2000 Lapita Interaction. Terra Australis 15. Canberra: Department of Archaeology and Natural

History and Centre for Archaeological Research, Australian National University.

SWINDLER, D. R., AND M. 1. WEISLER 2000 Dental size and morphology of precontact Marshall Islanders (Micronesia) compared with

other Pacific Islanders. Anthropological Science 108: 261-282.

TAKAYAMA,]. 1979 Archaeological investigations of P AAT -2 in the Palaus: An interim report, in Report: Cul­

tural Anthropological Research on the Folk Culture in the Western Caroline Islands of Micronesia in 1977: 81-103, ed. H. Kusakabe. Tokyo: Committee for Micronesian Research, Tokyo University of Foreign Studies.

TAKAYAMA,]., M. INTOH, AND H. TAKASUGI 1980 The Brief Archaeological Survey on Kayangel and Angaur in the Palaus. Reports of the Pacific

Archaeological Survey 7. Nara City: Tezukayama University.

TERRELL,]. E., AND R. L. WELSCH 1997 Lapita and the temporal geography of prehistory. Antiquity 71 : 548-572.

WATCHMAN, A. 1986 Brief petrological description of clay and pottery samples from the Siassi Islands and Huon

Peninsula, Papua New Guinea. Appendix 2 in Lilley 1986:512-532.

WEISLER, M. 2000 Burial artifacts from the Marshall Islands: Descriptions, dating and evidence for extra­

archipelago contacts. Micyonesica 33(1/2): 111-136.

WEISLER, M., AND D. SWINDLER 2002 Rocker jaws from the Marshall Islands: Evidence for interaction between eastern Micro­

nesia and West Polynesia. People and Culture in Oceania 18: 23-33.

WELSCH, R. L., AND]. E. TERRELL 1998 Material cnlture, social fields, and social boundaries on the Sepik coast of New Guinea, in

The Archaeology of Social Boundaries: 50-77, ed. M. T. Stark. Washington, DC: Smithso­nian Institution Press.

WICKLER, S. 2004 Modelling colonisation and migration in Micronesia from a zooarchaeological perspec­

tive, in Colonisation, Migration, and Marginal Areas: A Zooarchaeological Approach: 28-40, ed. M. Mondini, S. Munoz, and S. Wickler. Oxford: Oxbow Books.

WILSON, M. 2002 Picturing Pacific Prehistory: The Rock-Art of Vanuatu in a Western Pacific Context.

Unpubl. Ph.D. diss. Australian National University, Canberra.

WURM, S. A., AND S. HATTORI, EDS. 1981 Language Atlas of the Pacific Area. Canberra: Australian Academy of the Humanities, in col­

laboration with Japan Academy.

SPECHT ET AL. . TYPE X POTTERY AND MICRONESIAN RELATIONSHIPS 47

ABSTRACT

Type X is one of four Post-Lapita pottery styles reported from Huon Peninsula and the Siassi Islands of Papua New Guinea. Previous petrographic work was inconclu­sive about its likely area of origin but indicated a possible Huon Peninsula source. Renewed analysis of a larger sample supports this conclusion and confirms the use of grog temper. This kind of temper is otherwise not recorded in the New Guinea region, and its use in the production of Type X was probably culturally driven. Comparisons between Type X and grog-tempered pottery from Palau, Yap, and Pohnpei in Micronesia lead to the suggestion that Type X probably derived from an otherwise unrecorded contact between Huon Peninsula and Palau about 1000 years ago. The article reviews other evidence for interaction between the New Guinea-Bismarck Archipelago region and various parts of Micronesia and concludes that the proposed Type X connection with Palau is but one of several prehistoric contacts between different parts of the regions. Recognition of such contacts, which could have been unintentional and on a small scale, may contribute to explaining the complex ethnolinguistic situation of Huon Peninsula.